WITH VERFPLOETER WOUTER B. DE VRIES , RICARDO DE O. SCHMIDT, WES - - PowerPoint PPT Presentation

▶
with verfploeter
SMART_READER_LITE
LIVE PREVIEW

WITH VERFPLOETER WOUTER B. DE VRIES , RICARDO DE O. SCHMIDT, WES - - PowerPoint PPT Presentation

Apr 22, 2023 27 likes •437 views

BROAD AND LOAD-AWARE ANYCAST MAPPING WITH VERFPLOETER WOUTER B. DE VRIES , RICARDO DE O. SCHMIDT, WES HARDAKER, JOHN HEIDEMANN, PIETER-TJERK DE BOER AND AIKO PRAS London - November 3, 2017 ACM Internet Measurement Conference 2017 INTRODUCTION

slide-1
SLIDE 1

BROAD AND LOAD-AWARE ANYCAST MAPPING WITH VERFPLOETER

WOUTER B. DE VRIES, RICARDO DE O. SCHMIDT, WES HARDAKER, JOHN HEIDEMANN, PIETER-TJERK DE BOER AND AIKO PRAS London - November 3, 2017

slide-2
SLIDE 2

ACM Internet Measurement Conference 2017 2

INTRODUCTION

Anycast site User

slide-3
SLIDE 3

ACM Internet Measurement Conference 2017 2

INTRODUCTION

Users are routed to a (potentially) nearby site Anycast site User

slide-4
SLIDE 4

ACM Internet Measurement Conference 2017 2

INTRODUCTION

Users are routed to a (potentially) nearby site The set of clients that reach a site define that sites catchment Anycast site User

slide-5
SLIDE 5

ACM Internet Measurement Conference 2017 2

INTRODUCTION

Users are routed to a (potentially) nearby site The set of clients that reach a site define that sites catchment Catchments are often messy and not divided into neat geographic areas Anycast site User

slide-6
SLIDE 6

Peering agreements, Transits and BGP policies influence the routing process. You don’t know where a client from a certain prefix will be routed until he is routed there.

ACM Internet Measurement Conference 2017 3

INTRODUCTION

slide-7
SLIDE 7

Peering agreements, Transits and BGP policies influence the routing process. You don’t know where a client from a certain prefix will be routed until he is routed there.

ACM Internet Measurement Conference 2017 3

INTRODUCTION

Suboptimal routing can lead to severely degraded performance, in terms of RTT, load balancing and resilience

slide-8
SLIDE 8

ACM Internet Measurement Conference 2017 4

DETERMINING CATCHMENT - TODAY

  • Analyze log files
  • Only if you have them
  • Use active probing (e.g. RIPE Atlas)
  • Requires VPs at many locations
slide-9
SLIDE 9

ACM Internet Measurement Conference 2017 4

DETERMINING CATCHMENT - TODAY

  • Analyze log files
  • Only if you have them
  • Use active probing (e.g. RIPE Atlas)
  • Requires VPs at many locations

NL: 563 VPs

slide-10
SLIDE 10

ACM Internet Measurement Conference 2017 4

DETERMINING CATCHMENT - TODAY

  • Analyze log files
  • Only if you have them
  • Use active probing (e.g. RIPE Atlas)
  • Requires VPs at many locations

NL: 563 VPs China: 19 VPs Brazil: 63 VPs

slide-11
SLIDE 11
  • Develop a method to accurately map anycast

catchments

  • Without requiring input from users
  • “Calibrate” the anycast catchment to the actual client

base of the service

ACM Internet Measurement Conference 2017 5

GOALS

slide-12
SLIDE 12

ACM Internet Measurement Conference 2017 6

VERFPLOETER: METHODOLOGY

ACTIVE MEASUREMENT WITH “PASSIVE” VANTAGE POINTS Anycast sites Vantage points

slide-13
SLIDE 13

ACM Internet Measurement Conference 2017 6

VERFPLOETER: METHODOLOGY

ACTIVE MEASUREMENT WITH “PASSIVE” VANTAGE POINTS Send out ICMP Echo Requests Anycast sites Vantage points

slide-14
SLIDE 14

ACM Internet Measurement Conference 2017 6

VERFPLOETER: METHODOLOGY

ACTIVE MEASUREMENT WITH “PASSIVE” VANTAGE POINTS Anycast sites Vantage points ICMP Echo Replies are routed towards “nearest” anycast site

slide-15
SLIDE 15

ACM Internet Measurement Conference 2017 6

VERFPLOETER: METHODOLOGY

ACTIVE MEASUREMENT WITH “PASSIVE” VANTAGE POINTS Anycast sites Vantage points This defines the catchment that each VP belongs to

slide-16
SLIDE 16

ICMP-responding IPv4 or IPv6 addresses We use a recent IPv4 hitlist from USC/ISI: https://ant.isi.edu/datasets/ip_hitlists/ 1 IPv4 address per /24 block: 14.7 million addresses 6.3 million after removing those in which there was never a responding host

ACM Internet Measurement Conference 2017 7

WHAT ARE THE PASSIVE VPS

1.0.0.0/24 – 1.0.0.57 1.0.1.0/24 – 1.0.1.85 1.0.2.0/24 – 1.0.2.4 1.0.3.0/24 – 1.0.3.204

slide-17
SLIDE 17
  • The B Root DNS
  • Anycast since 1st of May 2017
  • 2 anycast sites
  • Our anycast testbed
  • 9 anycast sites

Measured catchment using both Atlas and Verfploeter

ACM Internet Measurement Conference 2017 8

USING VERFPLOETER: TWO CASE STUDIES

slide-18
SLIDE 18

ACM Internet Measurement Conference 2017 9

FROM ATLAS TO VERFPLOETER

B-ROOT CASE STUDY Atlas Verfploeter /24 blocks seen 8,677 (of 9083) 3,786,907 (of 6,877,175) Geolocatable 8,677 3,786,229 Unique 2,079 3,606,300 Geolocated using MaxMind GeoLite2

slide-19
SLIDE 19

ACM Internet Measurement Conference 2017 10

FROM ATLAS TO VERFPLOETER

B-ROOT CASE STUDY Note scale

slide-20
SLIDE 20

ACM Internet Measurement Conference 2017 10

FROM ATLAS TO VERFPLOETER

B-ROOT CASE STUDY Note scale Each “dot” is a pie chart

slide-21
SLIDE 21

Note scale

ACM Internet Measurement Conference 2017 11

FROM ATLAS TO VERFPLOETER

B-ROOT CASE STUDY Atlas scale

slide-22
SLIDE 22

Note scale

ACM Internet Measurement Conference 2017 11

FROM ATLAS TO VERFPLOETER

B-ROOT CASE STUDY Atlas Verfploeter Atlas scale

slide-23
SLIDE 23

Note scale

ACM Internet Measurement Conference 2017 11

FROM ATLAS TO VERFPLOETER

B-ROOT CASE STUDY Atlas Verfploeter Atlas scale

slide-24
SLIDE 24

Note scale

ACM Internet Measurement Conference 2017 11

FROM ATLAS TO VERFPLOETER

B-ROOT CASE STUDY Atlas Verfploeter Atlas scale

slide-25
SLIDE 25

A more diverse ground for verfploeter measurements: Our 9-site anycast testbed

ACM Internet Measurement Conference 2017 12

MEASURING AN ANYCAST TESTBED: TANGLED

Location Host Upstream AU Sydney Vultr AS20473 FR Paris Vultr AS20473 JP Tokyo WIDE AS2500 NL Enschede

  • Univ. of Twente

AS1103 UK London Vultr AS20473 US Washington USC/ISI AS1972 US Miami Florida Int. Univ. AS20080 BR Sao Paulo Florida Int. Univ. AS1251 DK Copenhagen DK-Hostmaster AS39389

slide-26
SLIDE 26

A more diverse ground for verfploeter measurements: Our 9-site anycast testbed

ACM Internet Measurement Conference 2017 12

MEASURING AN ANYCAST TESTBED: TANGLED

Location Host Upstream AU Sydney Vultr AS20473 FR Paris Vultr AS20473 JP Tokyo WIDE AS2500 NL Enschede

  • Univ. of Twente

AS1103 UK London Vultr AS20473 US Washington USC/ISI AS1972 US Miami Florida Int. Univ. AS20080 BR Sao Paulo Florida Int. Univ. AS1251 DK Copenhagen DK-Hostmaster AS39389 Location Host Upstream AU Sydney Vultr AS20473 FR Paris Vultr AS20473 JP Tokyo WIDE AS2500 NL Enschede

  • Univ. of Twente

AS1103 UK London Vultr AS20473 US Washington USC/ISI AS1972 US Miami Florida Int. Univ. AS20080 BR Sao Paulo Florida Int. Univ. AS1251 DK Copenhagen DK-Hostmaster AS39389

Shared upstream

slide-27
SLIDE 27

ACM Internet Measurement Conference 2017 13

MEASURING AN ANYCAST TESTBED: TANGLED

ATLAS VS VERFPLOETER

slide-28
SLIDE 28

ACM Internet Measurement Conference 2017 13

MEASURING AN ANYCAST TESTBED: TANGLED

ATLAS VS VERFPLOETER Atlas Verfploeter

slide-29
SLIDE 29

ACM Internet Measurement Conference 2017 13

MEASURING AN ANYCAST TESTBED: TANGLED

ATLAS VS VERFPLOETER Atlas Verfploeter

slide-30
SLIDE 30

Not all measured prefixes are equal Some will send no traffic, others a lot

ACM Internet Measurement Conference 2017 14

CALIBRATING CATCHMENT

e.g. due to NAT or ISP DNS Forwarders

slide-31
SLIDE 31

Prefix Site 1.0.0.0/24 LAX 1.0.1.0/24 LAX 1.0.2.0/24 LAX 1.0.3.0/24 MIA

ACM Internet Measurement Conference 2017 15

CALIBRATING CATCHMENT

Prefix Load 1.0.0.0/24 10 Q/s 1.0.1.0/24 0 Q/s 1.0.2.0/24 0 Q/s 1.0.3.0/24 15 Q/s RIGHT JOIN Obtained with Verfploeter Obtained from (historic) log files or from a service with a similar user base

slide-32
SLIDE 32

ACM Internet Measurement Conference 2017 15

CALIBRATING CATCHMENT

Prefix Site 1.0.0.0/24 LAX 1.0.1.0/24 LAX 1.0.2.0/24 LAX 1.0.3.0/24 MIA Prefix Load 1.0.0.0/24 10 Q/s 1.0.1.0/24 0 Q/s 1.0.2.0/24 0 Q/s 1.0.3.0/24 15 Q/s Prefix Site Load 1.0.0.0/24 LAX 10 Q/s 1.0.1.0/24 LAX 0 Q/s 1.0.2.0/24 LAX 0 Q/s 1.0.3.0/24 MIA 15 Q/s RIGHT JOIN

slide-33
SLIDE 33

ACM Internet Measurement Conference 2017 15

CALIBRATING CATCHMENT

Prefix Site 1.0.0.0/24 LAX 1.0.1.0/24 LAX 1.0.2.0/24 LAX 1.0.3.0/24 MIA Prefix Load 1.0.0.0/24 10 Q/s 1.0.1.0/24 0 Q/s 1.0.2.0/24 0 Q/s 1.0.3.0/24 15 Q/s Prefix Site Load 1.0.0.0/24 LAX 10 Q/s 1.0.1.0/24 LAX 0 Q/s 1.0.2.0/24 LAX 0 Q/s 1.0.3.0/24 MIA 15 Q/s RIGHT JOIN SITE Fraction of traffic LAX 0.40 MIA 0.60

slide-34
SLIDE 34
  • 1. Record traffic from unicast B-root (prior to anycast deployment)
  • 2. Run Verfploeter on anycast B-root

ACM Internet Measurement Conference 2017 16

CALIBRATING CATCHMENT AT B-ROOT

Date Site Queries/day Queries/sec 2017-04-12 LAX (Unicast) 2.34G 27.1k 2017-05-15 Both 2.20G 25.4k LAX 1.78G 20.6k MIA 0.407G 4.71k

slide-35
SLIDE 35

Date Method Measurement % to LAX May 15th RIPE Atlas 9682 VPs 82.4% May 15th Verfploeter 3.923M /24s 87.8% May 15th Verfploeter x Load 81.6% May 15th Actual load 2.188G q/day 81.4%

ACM Internet Measurement Conference 2017 17

CALIBRATING CATCHMENT AT B-ROOT

RECORDED TRAFFIC COMBINED WITH VERFPLOETER Before calibration After calibration

slide-36
SLIDE 36

ACM Internet Measurement Conference 2017 18

CALIBRATING CATCHMENT AT B-ROOT

Verfploeter allows you to experiment with various configurations and how that affects the load on the sites of your service

slide-37
SLIDE 37

ACM Internet Measurement Conference 2017 19

INSIDE AUTONOMOUS SYSTEMS

Anycast site User

slide-38
SLIDE 38

ACM Internet Measurement Conference 2017 19

INSIDE AUTONOMOUS SYSTEMS

Is one vantage point per AS enough?

Anycast site User

slide-39
SLIDE 39

ACM Internet Measurement Conference 2017 20

INSIDE AUTONOMOUS SYSTEMS

NUMBER OF SITES VERSUS NUMBER OF ANNOUNCED PREFIXES We see more than 1 site from around 12.7% of all ASes

slide-40
SLIDE 40

ACM Internet Measurement Conference 2017 21

CONCLUSIONS

  • Verfploeter has ~430x more “vantage points” than Atlas
  • Atlas has some blocks that we don’t have
  • Calibration of measured catchment to actual user base is required

Improve your anycasted service using Verfploeter https://github.com/woutifier/ Do your own analysis with our public datasets https://ant.isi.edu/datasets/anycast/index.html#verfploeter

slide-41
SLIDE 41

QUESTIONS AND COMMENTS

22

London - November 3, 2017